SAVITRIBAI PHULE PUNE UNIVERSITY

A PRELIMINARY PROJECT REPORT ON

Suspecius Activity

SUBMITTED TOWARDS THE

PARTIAL FULFILLMENT OF THE REQUIREMENTS OF

BACHELOR OF ENGINEERING (Computer

Engineering) BY

Student Name Exam No:

Student Name Exam No:
Student Name Exam No:
Student Name Exam No:

Under The Guidance of

Prof. Guide Name

DEPARTMENT OF COMPUTER ENGINEERING

Bharati Vidyapeeth’s College of Engineering for
Women Katraj-Dhankawadi
 Bharati Vidyapeeth’s College of Engineering for Women
DEPARTMENT OF COMPUTER ENGINEERING

CERTIFICATE

This is to certify that the Project

Entitled suspecious Activity

Detection Submitted by

Student Name Exam No:

Student Name Exam No:
Student Name Exam No:
Student Name Exam No:

is a bonafide work carried out by Students under the supervision of Prof. Guide
Name and it is submitted towards the partial fulfillment of the requirement of
Bach- elor of Engineering (Computer Engineering) Project.

Prof. Guide Name Prof. HOD Name

Internal Guide H.O.D
Dept. of Computer Engg. Dept. of Computer Engg.
 Abstract
suspecious Activity is predicting the body part or joint locations of
a person from an image or a video. This project will entail
detecting suspicious human Activity from real-time CCTV footage
using neu- ral networks. Human suspecious Activity is one of the
key problems in computer vision that has been studied for more
than 15 years. It is important because of the sheer number of
applications which can benefit from Activity detection. For
example, human pose esti- mation is used in applications including
video surveillance, animal tracking and behavior understanding,
sign language detection, ad- vanced human-computer interaction,
and marker less motion cap- turing. Low cost depth sensors have
limitations like limited to in- door use, and their low resolution and
noisy depth information make it difficult to estimate human poses
from depth images. Hence, we plan to use neural networks to
overcome these problems. Suspi- cious human activity recognition
from surveillance video is an active research area of image
processing and computer vision. Through the visual surveillance,
human activities can be monitored in sensi- tive and public areas
such as bus stations, railway stations, airports, banks, shopping
malls, school and colleges, parking lots, roads, etc. to prevent
terrorism, theft, accidents and illegal parking, vandalism, fighting,
chain snatching, crime and other suspicious activities. It is very
difficult to watch public places continuously, therefore an in-
telligent video surveillance is required that can monitor the human
activities in real-time and categorize them as usual and unusual ac-
tivities; and can generate an alert. Mostly, of the research being
carried out is on images and not videos. Also, none of the papers
published tries to use CNNs to detect suspicious activities

College Short Form Name, Department of Computer Engineering I

2021
 Acknowledgments

Please Write here Acknowledgment.Example given as

It gives us great pleasure in presenting the preliminary project report on ‘suspecious
Activity Detection’.

I would like to take this opportunity to thank my internal guide Prof. Guide Name
for giving me all the help and guidance I needed. I am really grateful to them for
their kind support. Their valuable suggestions were very helpful.

I am also grateful to Prof. HOD Name, Head of Computer Engineering Department,

CollegeName for his indispensable support, suggestions.

In the end our special thanks to Other Person Name for providing various
resources such as laboratory with all needed software platforms, continuous
Internet connec- tion, for Our Project.

Student Name1
Student Name2
Student Name3
Student Name4
(B.E. Computer Engg.)

College Short Form Name, Department of Computer Engineering II

2021
 INDEX

1 Introduction 1
1.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.2 Need . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.3 Objective . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2 Literature Survey 4
2.1 Study Of Research Paper . . . . . . . . . . . . . . . . . . . . . . . 5

3 Problem Statement 14

4 Project Requirement 16
4.1 EXTERNAL INTERFACE REQUIREMENT........................................17
4.1.1 User Interface..............................................................................17
4.1.2 Hardware Interfaces:....................................................................17
4.1.3 Software Interfaces......................................................................17
4.2 NON FUNCTIONAL REQUIREMENT.................................................18
4.2.1 Performance Requirements..........................................................18
4.2.2 Safety Requirement.....................................................................18
4.2.3 Software Quality Attributes.........................................................18

5 System Analysis 20
5.1 System Architecture.................................................................................21
5.1.1 Module.........................................................................................21
5.1.2 Data Flow Diagram.....................................................................21

5.2 UML DIAGRAMS...................................................................................22

6 Software Information 27
7 Project Plan 30
7.1 Stakeholder List . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
7.2 System Implementation Plan . . . . . . . . . . . . . . . . . . . . . 31

8 Conclusion 33
8.1 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

9 References 34

College Short Form Name, Department of Computer Engineering 2021 IV

List of Figures

5.1 System Architecture.................................................................................21

5.2 Data Flow diagram...................................................................................22
5.3 Data Flow diagram...................................................................................22
5.4 Data Flow diagram...................................................................................22
5.5 Use case Diagram.....................................................................................23
5.6 Activity Diagram......................................................................................24
5.7 Sequence Diagram....................................................................................25
5.8 Class Diagram..........................................................................................26
 CHAPTER 1

INTRODUCTION
1.1 OVERVIEW

We plan to build an application for detection of suspecious activity of people in

public places in real time. Our application can be used in surveillance at places like
malls, airports, railway stations, etc. where there is a risk of robbery or a shooting
attack. We will be using deep learning and neural networks to train our system.
This model will then be deployed as a mobile and desktop app which will take real
time CCTV footage as input and send an alert on the administrator’s device if
some suspicious pose is found. Human suspicious activity is related to identifying
human body parts and possibly tracking their movements. Real life applications of
it vary from gaming to AR/VR, to healthcare and gesture recognition. Compared to
image data domain, there is relatively little work on applying CNNs to video
classification. This is because, a video is more complex than images since it has
another dimension
- temporal. Unsupervised learning exploits temporal dependencies between frames
and has proven successful for video analysis. Some suspecious activity approaches
use CPU instead of GPU so that suspecious activity can run on low cost hardware
like embedded systems and mobile phones. Low cost depth sensors are another
new technology in computer vision. They are present in gaming consoles like the
Kinect for Xbox 360. They are motion sensors which allow the user to interact
with the console without a game controller, through just hand gestures. These are
RGB-D sensors that obtain depth information by structured light technology. The
structured light sensors infer the depth values by projecting an infrared light pattern
onto a scene and analyzing the distortion of the projected light pattern. However,
these sensors are limited to indoor use, and their low resolution and noisy depth
information make it difficult to estimate human poses from depth images.

1.1.1 Motivation

• Human suspecious activity is one of the key problems in computer vision

that has been studied for more than 15 years. It is important because of the
sheer number of applications which can benefit from suspecious activity. For
example, human suspecious activity is used in applications including video

College Short Form Name, Department of Computer Engineering 2

2021
 surveillance, animal tracking and behavior understanding, sign language de-
tection, advanced human-computer interaction, and marker less motion cap-
turing. Low cost depth sensors have limitations like limited to indoor use,
and their low resolution and noisy depth information make it difficult to es-
timate human poses from depth images. Hence, we plan to use neural net-
works to overcome these problems. Suspicious human activity recognition
from surveillance video is an active research area of image processing and
computer vision. Through the visual surveillance, human activities can be
monitored in sensitive and public areas such as bus stations, railway stations,
airports, banks, shopping malls, school and colleges, parking lots, roads, etc.
to prevent terrorism, theft, accidents and illegal parking, vandalism, fighting,
chain snatching, crime and other suspicious activities. It is very difficult to
watch public places continuously, therefore an intelligent video surveillance
is required that can monitor the human activities in real-time and categorize
them as usual and unusual activities; and can generate an alert. Mostly, of
the research being carried out is on images and not videos. Also, none of the
papers published tries to use CNNs to detect suspicious activities.

1.1.2 Need

• An crucial instrument in the battle against financial crimes like money launder-
ing is a suspicious activity report (SAR). All financial institutions are
advised to create a system for detecting and reporting suspicious activity by
the Finan- cial Action Task Force (FATF), a worldwide AML watchdog.

1.1.3 Objective

• Detect the suspecious activity

• Any one can use this application easy.

College Short Form Name, Department of Computer Engineering 3

2021
 CHAPTER 2

LITERATURE

SURVEY
2.1 STUDY OF RESEARCH PAPER

1. Paper Name:Real-Time suspecious Detection and Localization in Crowded Scenes

Author:Mohammad Sabokrou , Mahmood Fathy
Abstract :In this paper, we propose a method for real-time suspecious detection
and localization in crowded scenes. Each video is defined as a set of non-
overlapping cubic patches, and is described using two local and global descriptors.
These de- scriptors capture the video properties from different aspects. By
incorporating sim- ple and cost-effective Gaussian classifiers, we can distinguish
normal activities and anomalies in videos. The local and global features are based
on structure similarity between adjacent patches and the features learned in an
unsupervised way, using a sparse autoencoder. Experimental results show that our
algorithm is comparable to a state-of-the-art procedure on UCSD ped2 and UMN
benchmarks, but even more time-efficient. The experiments confirm that our
system can reliably detect and lo- calize anomalies as soon as they happen in a
video.

College Short Form Name, Department of Computer Engineering 5

2021
2. Paper Name:Learning Temporal Regularity in Video Sequences
Author:Mahmudul Hasan Jonghyun Choi
Abstract : : — Perceiving meaningful activities in a long video sequence is a chal-
lenging problem due to ambiguous definition of ‘meaningfulness’ as well as
clutters in the scene. We approach this problem by learning a generative model for
regu- lar motion patterns (termed as regularity) using multiple sources with very
limited supervision. Specifically, we propose two methods that are built upon the
autoen- coders for their ability to work with little to no supervision. We first
leverage the conventional handcrafted spatio-temporal local features and learn a
fully connected autoencoder on them. Second, we build a fully convolutional feed-
forward autoen- coder to learn both the local features and the classifiers as an end-
to-end learning framework. Our model can capture the regularities from multiple
datasets. We eval- uate our methods in both qualitative and quantitative ways -
showing the learned regularity of videos in various aspects and demonstrating
competitive performance on suspecious detection datasets as an application.

College Short Form Name, Department of Computer Engineering 6

2021
3. Paper Name:suspecious Detection in Video Using Predictive Convolutional
Long Short-Term Memory Networks
Author::Jefferson Ryan Medel
Description :—Automating the detection of anomalous events within long video
sequences is challenging due to the ambiguity of how such events are defined. We
approach the problem by learning generative models that can identify anomalies in
videos using limited supervision. We propose end-toend trainable composite Convo-
lutional Long Short-Term Memory (Conv-LSTM) networks that are able to predict
the evolution of a video sequence from a small number of input frames. Regular-
ity scores are derived from the reconstruction errors of a set of predictions with
abnormal video sequences yielding lower regularity scores as they diverge further
from the actual sequence over time. The models utilize a composite structure and
examine the effects of ‘conditioning’ in learning more meaningful representations.
The best model is chosen based on the reconstruction and prediction accuracies.
The Conv-LSTM models are evaluated both qualitatively and quantitatively,
demon- strating competitive results on suspecious detection datasets. Conv-LSTM
units are shown to be an effective tool for modeling and predicting video
sequences.

College Short Form Name, Department of Computer Engineering 7

2021
4. Paper Name:Abnormal Event Detection in Videos using Spatiotemporal
Autoen- coder
Author: : Yong Shean Chong
Description :—We present an efficient method for detecting anomalies in videos.
Recent applications of convolutional neural networks have shown promises of con-
volutional layers for object detection and recognition, especially in images. How-
ever, convolutional neural networks are supervised and require labels as learning
signals. We propose a spatiotemporal architecture for suspecious detection in videos
including crowded scenes. Our architecture includes two main components, one
for spatial feature representation, and one for learning the temporal evolution of the
spatial features. Experimental results on Avenue, Subway and UCSD benchmarks
confirm that the detection accuracy of our method is comparable to state-of-the-art
methods at a considerable speed of up to 140 fps.

College Short Form Name, Department of Computer Engineering 8

2021
5. Paper Name:Unrolled Optimization with Deep Priors
Author:Steven Diamond Vincent Sitzmann
Abstract:A broad class of problems at the core of computational imaging, sensing,
and low-level computer vision reduces to the inverse problem of extracting latent
images that follow a prior distribution, from measurements taken under a known
physical image formation model. Traditionally, handcrafted priors along with itera-
tive optimization methods have been used to solve such problems. In this paper we
present unrolled optimization with deep priors, a principled framework for infusing
knowledge of the image formation into deep networks that solve inverse problems
in imaging, inspired by classical iterative methods. We show that instances of the
framework outperform the state-of-the-art by a substantial margin for a wide
variety of imaging problems, such as denoising, deblurring, and compressed
sensing mag- netic resonance imaging (MRI). Moreover, we conduct experiments
that explain how the framework is best used and why it outperforms previous
methods.

College Short Form Name, Department of Computer Engineering 9

2021
6. paper Name:A Revisit of Sparse Coding Based suspecious Detection in Stacked
RNN Framework
Author:Weixin Luo
Abstract:—Motivated by the capability of sparse coding based suspecious detec-
tion, we propose a Temporally-coherent Sparse Coding (TSC) where we enforce
similar neighbouring frames be encoded with similar reconstruction coefficients.
Then we map the TSC with a special type of stacked Recurrent Neural Network
(sRNN). By taking advantage of sRNN in learning all parameters simultaneously,
the nontrivial hyper-parameter selection to TSC can be avoided, meanwhile with a
shallow sRNN, the reconstruction coefficients can be inferred within a forward
pass, which reduces the computational cost for learning sparse coefficients. The
contri- butions of this paper are two-fold: i) We propose a TSC, which can be
mapped to a sRNN which facilitates the parameter optimization and accelerates the
suspecious prediction. ii) We build a very large dataset which is even larger than
the summa- tion of all existing dataset for suspecious detection in terms of both the
volume of data and the diversity of scenes. Extensive experiments on both a toy
dataset and real datasets demonstrate that our TSC based and sRNN based method
consistently outperform existing methods, which validates the effectiveness of our
method

College Short Form Name, Department of Computer Engineering 10

2021
7. Paper Name:Connections Between Nuclear-Norm and FrobeniusNorm-Based
Representations
Author:Xi Peng, Canyi Lu, Zhang Y
Abstract:—A lot of works have shown that frobenius-norm-based representation
(FNR) is competitive to sparse representation and nuclearnorm-based
representation (NNR) in numerous tasks such as subspace clustering. Despite the
success of FNR in experimental studies, less theoretical analysis is provided to
understand its work- ing mechanism. In this brief, we fill this gap by building the
theoretical connections between FNR and NNR. More specially, we prove that: 1)
when the dictionary can provide enough representative capacity, FNR is exactly
NNR even though the data set contains the Gaussian noise, Laplacian noise, or
sample-specified corruption and
2) otherwise, FNR and NNR are two solutions on the column space of the dictionary.

College Short Form Name, Department of Computer Engineering 11

2021
8. Paper Name:A Review of Human suspecious activity from Single Image
Author:Naimat Ullah Khan, Wanggen Wan
Abstract:—This review is focused on the most significant contributions in Human
Pose Estimation methods from a single two-dimensional image. They start their
study with the traditional pictorial structure, go through a discussion of the use of
Deep Neural Networks that improved the human pose estimation significantly and
then the most recent, more famous approach namely Stacked Hourglass. Starting
from the first practical models for estimating human pose, they provide a compre-
hensive study of some of the most famous deep learning methods in order to
provide a concise analytical review of these most influential methods.

College Short Form Name, Department of Computer Engineering 12

2021
9. Paper name:Human Pose Estimation using Deep Structure Guided Learning
Author:Baole Ai, Yu Zhou, Yao Yu, Sidan Du Nanjing
Abstract:This paper presents an approach to incorporate structure knowledge into
CNNs for articulated human pose estimation from a single still image. Recent re-
search on pose estimation adopt CNNs as base blocks to combine with other graph-
ical models. Different from existing methods using features from CNNs to model
the tree structure, they directly use the structure pose prior to guide the learning of
CNN. First, they introduce a deep CNN with effective receptive fields which
capture the holistic context of the whole image. Second, limb loss is used as
intermediate supervision of CNN to learn the correlations of joints. Both parts and
joints features are extract the middle of neural network and then are used to guide
the following network learning.

College Short Form Name, Department of Computer Engineering 13

2021
 CHAPTER 3

PROBLEM STATEMENT
1. suspecious Activity is predicting the body part or joint locations of a person
from an image or a video. This project will entail detecting suspecious
human Activity from real-time CCTV footage using neural networks.

College Short Form Name, Department of Computer Engineering 15

2021
 CHAPTER 4

PROJECT REQUIREMENT
4.1 EXTERNAL INTERFACE REQUIREMENT

4.1.1 User Interface

• Application Based On suspecious Activity Detection.

4.1.2 Hardware Interfaces:

• Hardware : intel core

• Speed : 2.80 GHz

• RAM : 8GB

• HardDisk : 40 GB

• Key Board: Standard Windows Keyboard

4.1.3 Software Interfaces

• Operating System: Windows 10

• IDE: Annaconda

• Programming Language : Spyder

College Short Form Name, Department of Computer Engineering 17

2021
 4.2 NON FUNCTIONAL REQUIREMENT

4.2.1 Performance Requirements

• The performance of the functions and every module must be well. The
overall performance of the software will enable the users to work decently.
Perfor- mance of encryption of data should be fast. Performance of the
providing virtual environment should be fast Safety Requirement

• The application is designed in modules where errors can be detected and

steadily. This makes it easier to install and update new functionality if re-
quired.

4.2.2 Safety Requirement

• The application is designed in modules where errors can be detected and

fixed easily. This makes it easier to install and update new functionality if
required.

4.2.3 Software Quality Attributes

• Our software has many quality attribute that are given below:-

• Adaptability: This software is adaptable by all users.

• Availability: This software is freely available to all users. The availability of

the software is easy for everyone.

• Maintainability: After the deployment of the project if any error occurs then
it can be easily maintained by the software developer.

College Short Form Name, Department of Computer Engineering 18

2021
• Reliability: The performance of the software is better which will increase the
reliabilityof the Software.

• User Friendliness: Since, the software is a GUI application; the output gener-
ated is much user friendly in its behavior.

• Integrity: Integrity refers to the extent to which access to software or data by

unauthorized persons can be controlled.

• Security: Users are authenticated using many security phases so reliable

secu- rity is provided.

• Testability: The software will be tested considering all the aspects.

College Short Form Name, Department of Computer Engineering 19

2021
 CHAPTER 5

SYSTEM ANALYSIS
 5.1 SYSTEM ARCHITECTURE

Hardware Component

Raspberry Pi :
Raspberry Pi is defined as a minicomputer the size of a credit card that is interoperable with any
input and output hardware device like a monitor, a television, a mouse, or a keyboard – effectively
converting the set-up into a full-fledged PC at a low cost
Diagram

College Short Form Name, Department of Computer Engineering 21

2021
Features of Raspberry Pi
1. Central Processing Unit (CPU)
Every computer has a Central Processing Unit, and so does the Raspberry Pi. It is the computer’s brain
and carries out instructions using logical and mathematical operations. Raspberry Pi makes use of the
ARM11 series processor on its boards.
2. HDMI port
Raspberry Pi board has an HDMI or High Definition Multimedia Interface port that allows the device to
have video options of the output from the computer displayed. An HDMI cable connects the Raspberry Pi
to an HDTV. The supported versions include 1.3 and 1.3. It also comes with an RCA port for other
display options.
3. Graphic Processing Unit (GPU)
This unit, GPU or Graphic Processing Unit, is another part of the Raspberry pi board. Its primary purpose
is to hasten the speed of image calculations.
4. Memory (RAM)
Random Access Memory is a core part of a computer’s processing system. It is where real-time
information is stored for easy access. The initial Raspberry Pi had 256MB RAM. Over the years,
developers gradually and significantly improved the size. Different Raspberry Pi models come with
varying capacities. The model with the maximum capacity presently is the Raspberry Pi 4 with 8GB
RAM space.
5. Ethernet port
The Ethernet port is a connectivity hardware feature available on B models of Raspberry Pi. The Ethernet
port enables wired internet access to the minicomputer. Without it, software updates, web surfing, etc.,
would not be possible using the Raspberry Pi. The Ethernet port found on Raspberry computers uses the
RJ45 Ethernet jack. With this component, Raspberry Pi can connect to routers and other devices.
6. SD card slot
Like most other regular computers, Raspberry Pi must have some sort of storage device. However, unlike
conventional PCs, it does not come with a hard drive, nor does it come with a memory card. The
Raspberry Pi board has a Secure Digital card or SD card slot where users must insert SD cards for the
computer to function. The SD card functions like a hard drive as it contains the operating system
necessary for turning the system on. It also serves to store data.
7. General Purpose Input and Output (GPIO) pins
These are upward projecting pins in a cluster on one side of the board. The oldest models of the Raspberry
Pi had 26 pins, but most have 40 GPIO pins. These pins are pretty sensitive and should be handled
carefully. They are essential parts of the Raspberry Pi device as they add to its diverse applications. GPIO
College Short Form Name, Department of Computer Engineering 22
2021
pins are used to interact with other electronic circuits. They can read and control the electric signals from
other boards or devices based on how the user programs them.
8. LEDs
These are a group of five light-emitting diodes. They signal the user on the present status of the Raspberry
Pi unit. Their function covers:
PWR (Red): This functions solely to indicate power status. When the unit is on, it emits a red light and
only goes off when the unit is switched off, or disconnected from the power source.
ACT (Green): This flashes to indicate any form of SD card activity.
LNK (Orange): LNK LED gives off an orange light to signify that active Ethernet connectivity has been
established.
100 (Orange): This light comes on during Ethernet connection when the data speed reaches 100Mbps.
FDX (Orange): FDX light also comes during Ethernet connection. It shows that the connection is a full-
duplex.
9. USB ports
Universal service bus (USB) ports are a principal part of Raspberry Pi. They allow the computer to
connect to a keyboard, mouse, hard drives, etc. The first model of Raspberry Pi had only two USB 2.0
ports. Subsequent models increased this number to four. Raspberry Pi 4 and Pi 400, much newer models,
come with a mix of USB 2.0 and USB 3.0 ports.
10. Power source
Raspberry Pi has a power source connector that typically uses a 5V micro USB power cable. The amount
of electricity any Raspberry Pi consumes depends on what it’s used for and the number of peripheral
hardware devices connected.
How does Raspberry Pi work?
Raspberry Pi is a programmable device. It comes with all the critical features of the motherboard
in an average computer but without peripherals or internal storage. To set up the Raspberry computer, you
will need an SD card inserted into the provided space. The SD card should have the operating system
installed and is required for the computer to boot. Raspberry computers are compatible with Linux OS.
This reduces the amount of memory needed and creates an environment for diversity.
After setting up the OS, one can connect Raspberry Pi to output devices like computer monitors or
a High-Definition Multimedia Interface (HDMI) television. Input units like mice or keyboards should also
be connected. This minicomputer’s exact use and applications depend on the buyer and can cover many
functions.

o Operating System

College Short Form Name, Department of Computer Engineering 23

2021
• The Foundation provides Raspbian, a Debianbased Linux distribution for download, as well as third
party Ubuntu, Windows 10 IOT Core, RISC OS, and specialised media center distributions.
• It promotes Python and Scratch as the main programming language, with support for many other
languages.
• The default firmware is closed source, while an unofficial open source is available.
o Processor

• The Broadcom BCM2835 SoC used in the first generation Raspberry Pi is somewhat equivalent to the
chip used in first modern generation smartphones (its CPU is an older ARMv6 architecture), which
includes a 700 MHz ARM1176JZF-S processor, VideoCore IV graphics processing unit (GPU), and
RAM.
• It has a level 1 (L1) cache of 16 KB and a level 2 (L2) cache of 128 KB. The level 2 cache is used
primarily by the GPU. The SoC is stacked underneath the RAM chip, so only its edge is visible.
• The Raspberry Pi 2 uses a Broadcom BCM2836 SoC with a 900 MHz 32-bit quad-core ARM Cortex-A7
processor, with 256 KB shared L2 cache.
• The Raspberry Pi 3 uses a Broadcom BCM2837 SoC with a 1.2 GHz 64-bit quad-core ARM Cortex-
A53 processor, with 512 KB shared L2 cache
o Performance

• The Raspberry Pi 3, with a quad-core Cortex-A53 processor, is described as 10 times the performance of
a Raspberry Pi . This was suggested to be highly dependent upon task threading and instruction set use.
Benchmarks showed the Raspberry Pi 3 to be approximately 80% faster than the Raspberry Pi 2 in
parallelized tasks.
• Raspberry Pi 2 includes a quad-core Cortex-A7 CPU running at 900 MHz and 1 GB RAM. It is
described as 4–6 times more powerful than its predecessor. The GPU is identical to the original. In
parallelized benchmarks, the Raspberry Pi 2 could be up to 14 times faster than a Raspberry Pi 1 Model
B+.
o RAM

• The Raspberry Pi 2 and the Raspberry Pi 3 have 1 GB of RAM.

• The Raspberry Pi Zero and Zero W have 512 MB of RAM.
o Networking

• The Model A, A+ and Pi Zero have no Ethernet circuitry and are commonly connected to a network
using an external user-supplied USB Ethernet or Wi-Fi adapter.
• On the Model B and B+ the Ethernet port is provided by a built-in USB Ethernet adapter using the
SMSC LAN9514 chip.
• The Raspberry Pi 3 and Pi Zero W (wireless) are equipped with 2.4 GHz WiFi 802.11n (150 Mbit/s) and
Bluetooth 4.1 (24 Mbit/s) based on Broadcom BCM43438 Full MAC chip with no official support for
Monitor mode but implemented through unofficial firmware patching and the Pi 3 also has a 10/100
Ethernet port.

Peripherals
• The Raspberry Pi may be operated with any generic USB computer keyboard and mouse.
• It may also be used with USB storage, USB to MIDI converters, and virtually any other
device/component with USB capabilities.
• Other peripherals can be attached through the various pins and connectors on the surface of the
Raspberry Pi.
Video Capabilities
• The video controller can emit standard modern TV resolutions, such as HD and Full HD, and higher or
College Short Form Name, Department of Computer Engineering 24
2021
lower monitor resolutions and older standard CRT TV resolutions.
Uses of Raspberry Pi
1. Constructing a desktop PC
2. Enabling media usage
3. Controlling IoT robots
4. Acting as a printer server
5. Replacing web servers

Buzzer
Description:-
The Passive Buzzer Module KY-006 is a type of piezoelectric alarm sensor commonly used in DIY
electronics and robotics projects. It is a compact, self-contained module that generates an audible tone
when a voltage is applied to its input.
The KY-006 has a piezoelectric ceramic disc inside which vibrates when a voltage is applied, generating a
sound. It is passive, meaning that it does not have its own internal oscillator and must be driven by an
external signal.
The module has two pins, one for connecting to a positive voltage (VCC) and one for connecting to
ground (GND). When a square wave signal is applied to the input, the piezoelectric ceramic disc vibrates
at the frequency of the square wave, producing a sound.
The KY-006 is a small and simple module that can be easily integrated into a variety of projects and
devices. It can be used as a simple alarm, as a signal generator, or as a component in more complex audio
systems
There are two types of piezoelectric buzzers that are commonly used in electronics projects – active
buzzers and passive buzzers. Active buzzers are called active because they only need a DC voltage to
produce sound. Passive buzzers need an AC voltage to produce sound
Structural principle buzzer
(A) the introduction of the buzzer
1. The role of the buzzer: buzzer is an integrated structure of electronic transducers, DC power supply,
widely used in computers, printers, copiers, alarms, electronic toys, automotive electronic equipment,
telephones, timers, etc. electronic products for sound devices.
2. Buzzer Category: mainly divided into piezoelectric buzzer and two types of electromagnetic buzzer.
3. Buzzer circuit pattern symbols: the buzzer in the circuit by the letter "H" or "HA" (old standard with
"FM", "LB", "JD", etc.) indicates.
Schematic buzzer
(Two) structural principle buzzer
1. Piezo Buzzer: piezoelectric buzzer mainly by the multivibrator, piezo buzzer, impedance matching and
resonance box, casing and other components. Some piezo buzzer case is also equipped with light-emitting
diodes.
Multivibrator constituted by the transistors or integrated circuits. When switched on (1.5V-15V DC
working voltage), multi-harmonic oscillator start-up, the output 1.5-2.5kHZ audio signal, impedance
matching push piezo buzzer sound.
Piezo buzzer by a lead zirconate titanate or lead magnesium niobate piezoelectric ceramic material. Both
surfaces of the ceramic sheet plated silver electrode, the polarization and the aging treatment, and then
with brass or stainless steel sheet stick together.
2. Magnetic Buzzer: electromagnetic buzzer by the oscillator, the electromagnetic coil, magnet,
diaphragm and housing and other components.

Diagram:-

College Short Form Name, Department of Computer Engineering 25

2021
Features :-
Simple to use: The KY-006 has only two pins for power and ground, making it very easy to connect and
use.
Compact size: The small size of the KY-006 makes it suitable for use in portable and compact devices.
Low power consumption: The KY-006 consumes very little power, making it an ideal choice for battery-
powered devices.
Adjustable tone: The frequency and volume of the sound produced by the KY-006 can be adjusted by
changing the frequency and duty cycle of the input signal.
Wide operating voltage: The KY-006 can operate over a wide voltage range, typically between 3 and 5
volts, making it compatible with a variety of microcontrollers and control devices.
Easy to control: The KY-006 can be controlled by a microcontroller or other control device using digital
signals, making it easy to integrate into various projects and systems.
These features make the KY-006 an ideal choice for simple alarm systems, signal generators, and other
audio-related applications.
Pinout :-
VCC: Connects to a positive voltage source, typically between 3 and 5 volts.
GND: Connects to ground.
The KY-006 can be connected to a microcontroller or other control device using these two pins. To
generate a sound, a square wave signal is applied to the VCC pin, causing the piezoelectric ceramic disc
inside the KY-006 to vibrate at the frequency of the square wave, producing a sound.

Working:-
The KY-006 passive buzzer module is a piezoelectric alarm sensor that generates a tone when a voltage is
applied to it. It works by converting electrical energy into mechanical energy through the use of a
piezoelectric material, which vibrates at a specific frequency when voltage is applied. The module
produces an audible sound or tone, which can be used for alarms, signals, or notifications. To use the KY-
006 module, it needs to be connected to a microcontroller or a control circuit that can provide the
necessary voltage and control the frequency of the sound.

Specifications:-
Operating Voltage: DC 5V
Operating Current: 15mA
Operating Frequency: 2.048 kHz
Sound Pressure Level (SPL): 65±5 dB
Connector: 3-pin interface
Dimensions: 22mm x 15mm x 10mm

College Short Form Name, Department of Computer Engineering 26

2021
Weight: 2g
These specifications are approximate and may vary slightly between different manufacturers and models.
It is always a good idea to refer to the datasheet provided by the manufacturer for exact specifications and
to ensure compatibility with your project requirements.

Applications:-
Alarms: It can be used as a simple alarm sound generator, triggered by a microcontroller or a control
circuit.
Signaling: It can be used to provide audio feedback in response to user actions, such as button presses or
other events.
Notifications: It can be used to provide audio notifications, such as incoming message or call alerts in a
project.
Musical instrument: It can be used in a musical instrument to generate simple tones or beeps.
Educational projects: It can be used in educational projects to demonstrate the principles of
piezoelectricity and sound generation.
Type of sensor:-
The KY-006 Passive Buzzer Module is a type of piezoelectric alarm sensor. A piezoelectric sensor is a
device that converts electrical energy into mechanical energy and vice versa. In the case of the KY-006, it
uses a piezoelectric material to generate a sound or tone when a voltage is applied to it. The buzzer is
referred to as "passive" because it doesn't have an internal oscillator, and it requires an external signal to
generate the sound. This is in contrast to "active" buzzers, which have an internal oscillator and can
generate sound on their own.

5.1.1 Module

• Pre-processing

• Feature Extraction

• Classification

5.1.2 Data Flow Diagram

In Data Flow Diagram,we Show that flow of data in our system in DFD0 we show
that base DFD in which rectangle present input as well as output and circle show
our system,In DFD1 we show actual input and actual output of system input of our
College Short Form Name, Department of Computer Engineering 27
2021
system is text or image and output is rumor detected like wise in DFD 2 we present
operation of user as well as admin.

College Short Form Name, Department of Computer Engineering 28

2021
 Figure 5.2: Data Flow diagram

Figure 5.3: Data Flow diagram

Figure 5.4: Data Flow diagram

5.2 UML DIAGRAMS

Unified Modeling Language is a standard language for writing software blueprints.The

UML may be used to visualize,specify,construct and document the artifacts of a soft-
wareintensive system.UML is process independent,although optimally it should be
used in process that is use case driven,architecture-centric,iterative,and incremen-
tal.The Number of UML Diagram is available.

Use case Diagram.

Component Diagram.

College Short Form Name, Department of Computer Engineering 29

2021
Activity Diagram.

Sequence Diagram.

Figure 5.5: Use case Diagram

College Short Form Name, Department of Computer Engineering 30

2021
 Figure 5.6: Activity Diagram

College Short Form Name, Department of Computer Engineering 31

2021
 Figure 5.7: Sequence Diagram

College Short Form Name, Department of Computer Engineering 32

2021
 Figure 5.8: Class Diagram

College Short Form Name, Department of Computer Engineering 33

2021
 CHAPTER 6

SOFTWARE INFORMATION
Annaconda: What is Anaconda Navigator? Anaconda Navigator is a desktop
graphical user interface (GUI) included in Anaconda® distribution that allows you
to launch applications and easily manage conda packages, environments, and chan-
nels without using command-line commands. Navigator can search for packages
on Anaconda.org or in a local Anaconda Repository. It is available for Windows,
macOS, and Linux. To get Navigator, get the Navigator Cheat Sheet and install
Anaconda.
The Getting started with Navigator section shows how to start Navigator
from the shortcuts or from a terminal window.
What applications can I access using Navigator?
The following applications are available by default in Navigator:

1. JupyterLab

2. Jupyter Notebook

3. Spyder

4. PyCharm

5. VSCode

6. Glueviz

7. Orange 3 App

8. RStudio

College Short Form Name, Department of Computer Engineering 28

2021
 Spyder
Spyder is an open source cross-platform integrated development
environment (IDE) for scientific programming in the Python language. Spyder
integrates with a number of prominent packages in the scientific Python stack,
including NumPy, SciPy, Matplotlib, pandas, IPython, SymPy and Cython, as well
as other open source software.[3][4] It is released under the MIT license.[5]
Initially created and developed by Pierre Raybaut in 2009, since 2012 Spy-
der has been maintained and continuously improved by a team of scientific Python
developers and the community.
Spyder is extensible with first- and third-party plugins,[6] includes support
for interactive tools for data inspection and embeds Python-specific code quality
as- surance and introspection instruments, such as Pyflakes, Pylint[7] and Rope. It
is available cross-platform through Anaconda, on Windows, on macOS through
Mac- Ports, and on major Linux distributions such as Arch Linux, Debian, Fedora,
Gentoo Linux, openSUSE and Ubuntu.[8][9]
Spyder uses Qt for its GUI, and is designed to use either of the PyQt or
PySide Python bindings.[10] QtPy, a thin abstraction layer developed by the
Spyder project and later adopted by multiple other packages, provides the
flexibility to use either backend

College Short Form Name, Department of Computer Engineering 29

2021
 CHAPTER 7

PROJECT PLAN
In this chapter we are going to have an overview about how much time does it took
to complete each task like- Preliminray Survey Introduction and Problem
Statement, Literature Survey, Project Statement, Software Requirement and
Specification, Sys- tem Design, Partial Report Submission, Architecture Design,
Implementation, De- ployment, Testing, Paper Publish, Report Submission and
etcetera. This chapter also gives focus on stakeholder list which gives information
about project type, customer of the proposed system, user and project member who
developed the system.

7.1 STAKEHOLDER LIST

Sr. No. Stackholder

1 project Type
2. Customer
3 User

7.2 SYSTEM IMPLEMENTATION PLAN

The System Implementation plan table, shows the overall schedule of tasks compi-
lation and time duration required for each task.
Sr. No. Name/Title Start Date End Date
1 Preliminary Survey
2 Introduction and Problem State-
ment
3 Literature Survey
4 Project Statement
5 Software Requirement And Specifi-
cation
6 System Design
7 Partial Report Submission
8 Architecture Design
9 Implementation
10 Deployement
11 Testing
12 Paper Publish
13 Report Submission
Chapter 8

Conclusion

8.1 CONCLUSION

A system to process real-time CCTV footage to detect any suspecious activity will
help to create better security and less human intervention.Great strides have been
made in the field of human suspecious Activity, which enables us to better serve
the myriad applications that are possible with it. Moreover, research in related
fields such as Activity Tracking can greatly enhance its productive utilization in
several fields.
Chapter 9

References

1 Eralda Nishani, Betim Cico : “Computer Vision Approaches based on Deep

Learning and Neural Networks” Deep Neural Networks for Video Analysis
of Human Pose Estimation- 2017 6th MEDITERRANEAN CONFERENCE
ON EMBEDDED COMPUTING (MECO), 11-15 JUNE 2017, BAR, MON-
TENEGRO

2 Naimat Ullah Khan , Wanggen Wan : “A Review of Human Pose Estimation

from Single Image”- 978-1-5386-5195-7/18/ 2018 IEEE

3 Qiuhui Chen, Chongyang Zhang, Weiwei Liu, and Dan Wang, ”Surveillance
Human Pose Dataset And Performance Evaluation For Coarse-Grained Pose
Estimation”, Athens 2018.

4 Baole Ai, Yu Zhou, Yao Yu : “Human Pose Estimation using Deep Structure
Guided Learning”- 978-1-5090-4822-9/17 2017 IEEE DOI 10.1109/WACV.2017.141

5 Zhe Cao, Tomas Simon, Shih-En Wei, Yaser Sheikh The Robotics Institute,
Carnegie Mellon University ”Real time Multiperson 2D Pose Estimation
using
 part affinity fields” - 1063-6919/17 2017 IEEE DOI 10.1109/CVPR.2017.143

6 Hanguen Kim, Sangwon Lee, Dongsung Lee, Soonmin Choi, JinsunJu and
Huyun Myung “Real-Time Human Pose Estimation and Gesture Recognition
from depth Images Using Superpixels and SVM classifier.”- Sensors 2015,
15, 12410-12427; doi:10.3390/s150612410

7 Tripathi, Rajesh and Jalal, Anand and Agarwal, Subhash(2017). ”Suspicious

Human Activity Recognition: a Review”. Artificial Intelligence Review. 50.10.1007/s10462-
017-9545-7.

8 E. Eksioglu. Decoupled algorithm for MRI reconstruction using nonlocal

block matching model: BM3DMRI. Journal of Mathematical Imaging and
Vision, 56(3):430–440, 2016.

9 S. Wang, Z. Su, L. Ying, X. Peng, S. Zhu, F. Liang, D. Feng, and D. Liang. Ac-
celerating magnetic resonance imaging via deep learning. In Proceedings of
the IEEE International Symposium on Biomedical Imaging, pages 514–517,
2016.

10 L. Xu, J. Ren, C. Liu, and J. Jia. Deep convolutional neural network for
image deconvolution. In Advances in Neural Information Processing
Systems, pages 1790–1798, 2014.

11 Y. Yang, J. Sun, H. Li, and Z. Xu. Deep ADMM-Net for compressive sensing
MRI. In Advances in Neural Information Processing Systems, pages 10–18,
2016.

12 Z. Zhan, J.-F. Cai, D. Guo, Y. Liu, Z. Chen, and X. Qu. Fast multiclass dic-
tionaries learning with geometrical directions in MRI reconstruction. IEEE
Transactions on Biomedical Engineering, 63(9):1850– 1861, 2016.